Inspecting the climate emotions data

Descriptives

vars n mean sd median trimmed mad min max range skew kurtosis se
climate.anger 1 239 3.849372 0.9400684 4.00 3.950777 1.11195 1 5 4 -0.8236903 0.0927332 0.0608080
climate.contempt 2 239 2.407950 1.0096174 2.25 2.323834 1.11195 1 5 4 0.6608964 -0.0304711 0.0653067
climate.enthusiasm 3 239 3.599372 0.8169101 3.75 3.658031 0.74130 1 5 4 -0.7574711 0.4252301 0.0528415
climate.powerlessness 4 239 3.283473 0.7771209 3.25 3.323834 0.74130 1 5 4 -0.5012505 0.2529005 0.0502678
climate.guilt 5 239 2.921548 0.9676306 3.00 2.965026 1.11195 1 5 4 -0.2744229 -0.6667040 0.0625908
climate.isolation 6 239 2.821130 0.9430547 3.00 2.818653 1.11195 1 5 4 0.0587578 -0.6621222 0.0610011
climate.anxiety 7 239 3.610879 0.9403127 4.00 3.693005 0.74130 1 5 4 -0.8687992 0.1356662 0.0608238
climate.sorrow 8 239 4.064854 0.8417164 4.25 4.183938 0.74130 1 5 4 -1.2282603 1.3711738 0.0544461

Distributions

## $climate.anger
## 
##  Shapiro-Wilk normality test
## 
## data:  X[[i]]
## W = 0.92004, p-value = 0.0000000004786
## 
## 
## $climate.contempt
## 
##  Shapiro-Wilk normality test
## 
## data:  X[[i]]
## W = 0.94316, p-value = 0.00000005074
## 
## 
## $climate.enthusiasm
## 
##  Shapiro-Wilk normality test
## 
## data:  X[[i]]
## W = 0.94184, p-value = 0.00000003779
## 
## 
## $climate.powerlessness
## 
##  Shapiro-Wilk normality test
## 
## data:  X[[i]]
## W = 0.96874, p-value = 0.00004144
## 
## 
## $climate.guilt
## 
##  Shapiro-Wilk normality test
## 
## data:  X[[i]]
## W = 0.96677, p-value = 0.0000227
## 
## 
## $climate.isolation
## 
##  Shapiro-Wilk normality test
## 
## data:  X[[i]]
## W = 0.98014, p-value = 0.001968
## 
## 
## $climate.anxiety
## 
##  Shapiro-Wilk normality test
## 
## data:  X[[i]]
## W = 0.91516, p-value = 0.0000000002001
## 
## 
## $climate.sorrow
## 
##  Shapiro-Wilk normality test
## 
## data:  X[[i]]
## W = 0.88163, p-value = 0.000000000001055

Internal consistencies

We inspect the internal consistencies of the ICE scales using Cronbach’s alpha coefficient

anger contempt enthusiasm powerlessness guilt isolation anxiety sorrow
raw_alpha 0.892050814049292 0.857519620165384 0.858242392709822 0.679712971588263 0.882768915185896 0.845797968067275 0.899176100707087 0.899513338978477
std.alpha 0.890417086788687 0.857609797012417 0.858336050474642 0.67831814854768 0.881891491107175 0.845717534193633 0.899190341256113 0.899953324857259

Inspecting the Mental Health Continuum data

Descriptives

vars n mean sd median trimmed mad min max range skew kurtosis se
emo_wb 1 239 3.948396 1.237689 4.0 4.008636 1.48260 1.000000 6 5.000000 -0.4163687 -0.6932443 0.0800594
soc_wb 2 239 3.195816 1.153649 3.2 3.180311 1.18608 1.000000 6 5.000000 0.0566383 -0.7028915 0.0746234
psy_wb 3 239 3.992329 1.061842 4.0 4.034542 0.98840 1.166667 6 4.833333 -0.3889437 -0.3331321 0.0686849

Distributions

## $emo_wb
## 
##  Shapiro-Wilk normality test
## 
## data:  X[[i]]
## W = 0.96002, p-value = 0.000003284
## 
## 
## $soc_wb
## 
##  Shapiro-Wilk normality test
## 
## data:  X[[i]]
## W = 0.98059, p-value = 0.002326
## 
## 
## $psy_wb
## 
##  Shapiro-Wilk normality test
## 
## data:  X[[i]]
## W = 0.97851, p-value = 0.001083

Internal consistencies

We inspect the internal consistencies of the MHC scales using Cronbach’s alpha coefficient

emo_wb soc_wb psy_wb
raw_alpha 0.855735096094993 0.846650878994999 0.860443041538284
std.alpha 0.855753354854176 0.848789196268397 0.860883174998539

Demographics

Descriptives

vars n mean sd median trimmed mad min max range skew kurtosis se
cc_concern 1 239 3.4225941 1.0578200 4 3.4870466 1.4826 1 5 4 -0.5608707 -0.1254629 0.0684247
age 2 239 46.5690377 15.8271401 46 46.4974093 20.7564 19 74 55 0.0404981 -1.2454797 1.0237726
gender 3 239 0.5355649 0.4997802 1 0.5440415 0.0000 0 1 1 -0.1417265 -1.9881799 0.0323281

Distributions

## 
##  Shapiro-Wilk normality test
## 
## data:  demographics$cc_concern
## W = 0.88709, p-value = 0.000000000002299
## 
##  Shapiro-Wilk normality test
## 
## data:  demographics$age
## W = 0.95016, p-value = 0.0000002598

Climate emotions & mental health

Correlation table

Let’s use the Spearman correlation coefficient because the data is ordinal and non-parametrically distributed

climate.anger climate.contempt climate.enthusiasm climate.powerlessness climate.guilt climate.isolation climate.anxiety climate.sorrow emo_wb soc_wb psy_wb cc_concern age gender
climate.anger
climate.contempt -0.63***
climate.enthusiasm 0.44*** -0.38***
climate.powerlessness 0.33*** -0.16* 0.16*
climate.guilt 0.46*** -0.40*** 0.37*** 0.53***
climate.isolation 0.47*** -0.18** 0.37*** 0.41*** 0.49***
climate.anxiety 0.72*** -0.57*** 0.48*** 0.44*** 0.60*** 0.55***
climate.sorrow 0.75*** -0.62*** 0.42*** 0.34*** 0.45*** 0.44*** 0.74***
emo_wb 0.12 -0.10 0.25*** 0.01 0.04 0.05 0.17** 0.10
soc_wb 0.16* -0.15* 0.40*** 0.09 0.15* 0.22*** 0.27*** 0.15* 0.64***
psy_wb 0.18** -0.14* 0.35*** 0.02 0.02 0.12 0.19** 0.17** 0.78*** 0.76***
cc_concern 0.74*** -0.63*** 0.53*** 0.31*** 0.60*** 0.49*** 0.76*** 0.72*** 0.17** 0.27*** 0.24***
age 0.07 0.01 0.13 0.05 0.03 0.21** 0.02 0.14* 0.19** 0.27*** 0.31*** 0.10
gender 0.03 0.08 -0.08 -0.13* -0.29*** -0.08 -0.08 -0.12 0.02 0.06 0.06 -0.09 0.03

Climate emotions as predictors of mental wellbeing

Emotional wellbeing

The multiple linear regression model with climate enthusiasm, climate anxiety, age, and climate concern as predictors

## 
## Call:
## lm(formula = emo_wb ~ climate.enthusiasm + climate.anxiety + 
##     age + cc_concern, data = df_wb)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -2.9023 -0.7704  0.1735  0.8760  2.5495 
## 
## Coefficients:
##                    Estimate Std. Error t value   Pr(>|t|)    
## (Intercept)        2.099623   0.421550   4.981 0.00000123 ***
## climate.enthusiasm 0.240779   0.120580   1.997     0.0470 *  
## climate.anxiety    0.060717   0.142655   0.426     0.6708    
## age                0.012700   0.004947   2.568     0.0109 *  
## cc_concern         0.050088   0.126989   0.394     0.6936    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.195 on 234 degrees of freedom
## Multiple R-squared:  0.08286,    Adjusted R-squared:  0.06718 
## F-statistic: 5.285 on 4 and 234 DF,  p-value: 0.0004307

Standardised regressions coefficients

## 
## Call:
## lm(formula = emo_wb ~ climate.enthusiasm + climate.anxiety + 
##     age + cc_concern, data = df_wb)
## 
## Standardized Coefficients::
##        (Intercept) climate.enthusiasm    climate.anxiety                age         cc_concern 
##                 NA         0.15892076         0.04612844         0.16240951         0.04280899

Squared Semi-partial correlation coefficients

## Predictor 1: semi partial = 0.126; squared semipartial = 0.016
## Predictor 2: semi partial = 0.032; squared semipartial = 0.001
## Predictor 3: semi partial = 0.161; squared semipartial = 0.026
## Predictor 4: semi partial = 0.032; squared semipartial = 0.001

95% bootstrapped and accelerated confidence intervals of the regression coefficients

For the intercept

## BOOTSTRAP CONFIDENCE INTERVAL CALCULATIONS
## Based on 1000 bootstrap replicates
## 
## CALL : 
## boot.ci(boot.out = results, type = "bca", index = 1)
## 
## Intervals : 
## Level       BCa          
## 95%   ( 1.337,  2.928 )  
## Calculations and Intervals on Original Scale

For climate enthusiasm

## BOOTSTRAP CONFIDENCE INTERVAL CALCULATIONS
## Based on 1000 bootstrap replicates
## 
## CALL : 
## boot.ci(boot.out = results, type = "bca", index = 2)
## 
## Intervals : 
## Level       BCa          
## 95%   ( 0.0375,  0.5061 )  
## Calculations and Intervals on Original Scale

For climate anxiety

## BOOTSTRAP CONFIDENCE INTERVAL CALCULATIONS
## Based on 1000 bootstrap replicates
## 
## CALL : 
## boot.ci(boot.out = results, type = "bca", index = 3)
## 
## Intervals : 
## Level       BCa          
## 95%   (-0.2191,  0.3631 )  
## Calculations and Intervals on Original Scale

For age

## BOOTSTRAP CONFIDENCE INTERVAL CALCULATIONS
## Based on 1000 bootstrap replicates
## 
## CALL : 
## boot.ci(boot.out = results, type = "bca", index = 4)
## 
## Intervals : 
## Level       BCa          
## 95%   ( 0.0026,  0.0224 )  
## Calculations and Intervals on Original Scale

For climate concern

## BOOTSTRAP CONFIDENCE INTERVAL CALCULATIONS
## Based on 1000 bootstrap replicates
## 
## CALL : 
## boot.ci(boot.out = results, type = "bca", index = 5)
## 
## Intervals : 
## Level       BCa          
## 95%   (-0.2284,  0.3673 )  
## Calculations and Intervals on Original Scale

Assumptions check

1. Distribution of the model residuals

Let’s have a look at the plots of model residuals:

2. Linear relationship between independent and dependent variables

3. No multicollinearity

We know from the correlation matrix that correlations of the independent variables do not exceed the customary cutoff point of .8 so we can say that this assumption is met. But let’s also have a look at the Variance Inflation Factor:

## climate.enthusiasm    climate.anxiety                age         cc_concern 
##           1.616067           2.996910           1.020857           3.005494

The value for VIF starts at 1 and has no upper limit. A general rule of thumb for interpreting VIFs is as follows:

  • A value of 1 indicates there is no correlation between a given predictor variable and any other predictor variables in the model.
  • A value between 1 and 5 indicates moderate correlation between a given predictor variable and other predictor variables in the model, but this is often not severe enough to require attention.
  • A value greater than 5 indicates potentially severe correlation between a given predictor variable and other predictor variables in the model. In this case, the coefficient estimates and p-values in the regression output are likely unreliable.
4. Homoscedasticity

Let’s check the plot of the predicted values against the standardized residual values from point 1 to confirm that the points are equally distributed across all the values of the independent variables.

Social wellbeing

The multiple linear regression model with climate anger, climate contempt, climate enthusiasm, climate guilt, climate isolation, climate anxiety, climate sorrow , age, and climate concern as predictors

## 
## Call:
## lm(formula = soc_wb ~ climate.anger + climate.contempt + climate.enthusiasm + 
##     climate.guilt + climate.isolation + climate.anxiety + climate.sorrow + 
##     age + cc_concern, data = df_wb)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -2.42571 -0.66727 -0.00396  0.71477  2.99912 
## 
## Coefficients:
##                     Estimate Std. Error t value   Pr(>|t|)    
## (Intercept)         0.675660   0.717356   0.942   0.347249    
## climate.anger      -0.132380   0.131247  -1.009   0.314214    
## climate.contempt    0.014965   0.106567   0.140   0.888446    
## climate.enthusiasm  0.509901   0.103861   4.909 0.00000174 ***
## climate.guilt      -0.128479   0.096746  -1.328   0.185498    
## climate.isolation  -0.012636   0.095521  -0.132   0.894876    
## climate.anxiety     0.286214   0.143716   1.992   0.047612 *  
## climate.sorrow     -0.175979   0.145262  -1.211   0.226968    
## age                 0.016664   0.004362   3.820   0.000172 ***
## cc_concern          0.138841   0.127961   1.085   0.279053    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.02 on 229 degrees of freedom
## Multiple R-squared:  0.248,  Adjusted R-squared:  0.2184 
## F-statistic: 8.391 on 9 and 229 DF,  p-value: 0.00000000008279

Standardised regressions coefficients

## 
## Call:
## lm(formula = soc_wb ~ climate.anger + climate.contempt + climate.enthusiasm + 
##     climate.guilt + climate.isolation + climate.anxiety + climate.sorrow + 
##     age + cc_concern, data = df_wb)
## 
## Standardized Coefficients::
##        (Intercept)      climate.anger   climate.contempt climate.enthusiasm      climate.guilt  climate.isolation 
##                 NA        -0.10787211         0.01309646         0.36106584        -0.10776241        -0.01032931 
##    climate.anxiety     climate.sorrow                age         cc_concern 
##         0.23328662        -0.12839630         0.22861792         0.12730794

Squared Semi-partial correlation coefficients

## Predictor 1: semi partial = 0.055; squared semipartial = 0.003
## Predictor 2: semi partial = 0; squared semipartial = 0
## Predictor 3: semi partial = 0.281; squared semipartial = 0.079
## Predictor 4: semi partial = 0.077; squared semipartial = 0.006
## Predictor 5: semi partial = 0; squared semipartial = 0
## Predictor 6: semi partial = 0.114; squared semipartial = 0.013
## Predictor 7: semi partial = 0.071; squared semipartial = 0.005
## Predictor 8: semi partial = 0.219; squared semipartial = 0.048
## Predictor 9: semi partial = 0.063; squared semipartial = 0.004

Plotting the relationship between social wellbeing, enthusiasm and anxiety

95% bootstrapped and accelerated confidence intervals of the regression coefficients

For the intercept

## BOOTSTRAP CONFIDENCE INTERVAL CALCULATIONS
## Based on 1000 bootstrap replicates
## 
## CALL : 
## boot.ci(boot.out = results, type = "bca", index = 1)
## 
## Intervals : 
## Level       BCa          
## 95%   (-0.9595,  2.2224 )  
## Calculations and Intervals on Original Scale

For climate anger

## BOOTSTRAP CONFIDENCE INTERVAL CALCULATIONS
## Based on 1000 bootstrap replicates
## 
## CALL : 
## boot.ci(boot.out = results, type = "bca", index = 2)
## 
## Intervals : 
## Level       BCa          
## 95%   (-0.4748,  0.1369 )  
## Calculations and Intervals on Original Scale

For climate contempt

## BOOTSTRAP CONFIDENCE INTERVAL CALCULATIONS
## Based on 1000 bootstrap replicates
## 
## CALL : 
## boot.ci(boot.out = results, type = "bca", index = 3)
## 
## Intervals : 
## Level       BCa          
## 95%   (-0.2304,  0.2608 )  
## Calculations and Intervals on Original Scale

For climate enthusiasm

## BOOTSTRAP CONFIDENCE INTERVAL CALCULATIONS
## Based on 1000 bootstrap replicates
## 
## CALL : 
## boot.ci(boot.out = results, type = "bca", index = 4)
## 
## Intervals : 
## Level       BCa          
## 95%   ( 0.2884,  0.7059 )  
## Calculations and Intervals on Original Scale

For climate guilt

## BOOTSTRAP CONFIDENCE INTERVAL CALCULATIONS
## Based on 1000 bootstrap replicates
## 
## CALL : 
## boot.ci(boot.out = results, type = "bca", index = 5)
## 
## Intervals : 
## Level       BCa          
## 95%   (-0.3152,  0.0763 )  
## Calculations and Intervals on Original Scale

For climate isolation

## BOOTSTRAP CONFIDENCE INTERVAL CALCULATIONS
## Based on 1000 bootstrap replicates
## 
## CALL : 
## boot.ci(boot.out = results, type = "bca", index = 6)
## 
## Intervals : 
## Level       BCa          
## 95%   (-0.2196,  0.1619 )  
## Calculations and Intervals on Original Scale

For climate anxiety

## BOOTSTRAP CONFIDENCE INTERVAL CALCULATIONS
## Based on 1000 bootstrap replicates
## 
## CALL : 
## boot.ci(boot.out = results, type = "bca", index = 7)
## 
## Intervals : 
## Level       BCa          
## 95%   (-0.0057,  0.5834 )  
## Calculations and Intervals on Original Scale

For climate sorrow

## BOOTSTRAP CONFIDENCE INTERVAL CALCULATIONS
## Based on 1000 bootstrap replicates
## 
## CALL : 
## boot.ci(boot.out = results, type = "bca", index = 8)
## 
## Intervals : 
## Level       BCa          
## 95%   (-0.4597,  0.1573 )  
## Calculations and Intervals on Original Scale

For age

## BOOTSTRAP CONFIDENCE INTERVAL CALCULATIONS
## Based on 1000 bootstrap replicates
## 
## CALL : 
## boot.ci(boot.out = results, type = "bca", index = 9)
## 
## Intervals : 
## Level       BCa          
## 95%   ( 0.0082,  0.0254 )  
## Calculations and Intervals on Original Scale

For climate concern

## BOOTSTRAP CONFIDENCE INTERVAL CALCULATIONS
## Based on 1000 bootstrap replicates
## 
## CALL : 
## boot.ci(boot.out = results, type = "bca", index = 10)
## 
## Intervals : 
## Level       BCa          
## 95%   (-0.1536,  0.3888 )  
## Calculations and Intervals on Original Scale

Assumptions check

1. Distribution of the model residuals

Let’s have a look at the plots of model residuals:

2. Linear relationship between independent and dependent variables

3. No multicollinearity

We know from the correlation matrix that correlations of the independent variables do not exceed the customary cutoff point of .8 so we can say that this assumption is met. But let’s also have a look at the Variance Inflation Factor:

##      climate.anger   climate.contempt climate.enthusiasm      climate.guilt  climate.isolation    climate.anxiety 
##           3.483100           2.648661           1.647096           2.005172           1.856709           4.178555 
##     climate.sorrow                age         cc_concern 
##           3.420610           1.090562           4.192281

The value for VIF starts at 1 and has no upper limit. A general rule of thumb for interpreting VIFs is as follows:

  • A value of 1 indicates there is no correlation between a given predictor variable and any other predictor variables in the model.
  • A value between 1 and 5 indicates moderate correlation between a given predictor variable and other predictor variables in the model, but this is often not severe enough to require attention.
  • A value greater than 5 indicates potentially severe correlation between a given predictor variable and other predictor variables in the model. In this case, the coefficient estimates and p-values in the regression output are likely unreliable.
4. Homoscedasticity

Let’s check the plot of the predicted values against the standardized residual values from point 1 to confirm that the points are equally distributed across all the values of the independent variables.

Psychological wellbeing

The multiple linear regression model with climate anger, climate contempt, climate enthusiasm, climate anxiety, climate sorrow, age, and climate concern as predictors

## 
## Call:
## lm(formula = psy_wb ~ climate.anger + climate.contempt + climate.enthusiasm + 
##     climate.anxiety + climate.sorrow + age + cc_concern, data = df_wb)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -2.88287 -0.55304  0.07385  0.63997  2.61958 
## 
## Coefficients:
##                     Estimate Std. Error t value  Pr(>|t|)    
## (Intercept)         1.807346   0.663710   2.723  0.006961 ** 
## climate.anger       0.041766   0.123242   0.339  0.734998    
## climate.contempt    0.003068   0.096898   0.032  0.974772    
## climate.enthusiasm  0.375773   0.099179   3.789  0.000193 ***
## climate.anxiety    -0.017045   0.127154  -0.134  0.893477    
## climate.sorrow     -0.044280   0.139507  -0.317  0.751226    
## age                 0.017670   0.004119   4.290 0.0000262 ***
## cc_concern          0.024231   0.120680   0.201  0.841041    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.9797 on 231 degrees of freedom
## Multiple R-squared:  0.1737, Adjusted R-squared:  0.1487 
## F-statistic: 6.938 on 7 and 231 DF,  p-value: 0.0000001675

Standardised regressions coefficients

## 
## Call:
## lm(formula = psy_wb ~ climate.anger + climate.contempt + climate.enthusiasm + 
##     climate.anxiety + climate.sorrow + age + cc_concern, data = df_wb)
## 
## Standardized Coefficients::
##        (Intercept)      climate.anger   climate.contempt climate.enthusiasm    climate.anxiety     climate.sorrow 
##                 NA        0.036976222        0.002916741        0.289094390       -0.015094541       -0.035100284 
##                age         cc_concern 
##        0.263380391        0.024139296

Squared Semi-partial correlation coefficients

## Predictor 1: semi partial = 0; squared semipartial = 0
## Predictor 2: semi partial = 0; squared semipartial = 0
## Predictor 3: semi partial = 0.226; squared semipartial = 0.051
## Predictor 4: semi partial = 0; squared semipartial = 0
## Predictor 5: semi partial = 0; squared semipartial = 0
## Predictor 6: semi partial = 0.257; squared semipartial = 0.066
## Predictor 7: semi partial = 0; squared semipartial = 0

95% bootstrapped and accelerated confidence intervals of the regression coefficients

For the intercept

## BOOTSTRAP CONFIDENCE INTERVAL CALCULATIONS
## Based on 1000 bootstrap replicates
## 
## CALL : 
## boot.ci(boot.out = results, type = "bca", index = 1)
## 
## Intervals : 
## Level       BCa          
## 95%   ( 0.443,  3.290 )  
## Calculations and Intervals on Original Scale

For climate anger

## BOOTSTRAP CONFIDENCE INTERVAL CALCULATIONS
## Based on 1000 bootstrap replicates
## 
## CALL : 
## boot.ci(boot.out = results, type = "bca", index = 2)
## 
## Intervals : 
## Level       BCa          
## 95%   (-0.2301,  0.3596 )  
## Calculations and Intervals on Original Scale

For climate contempt

## BOOTSTRAP CONFIDENCE INTERVAL CALCULATIONS
## Based on 1000 bootstrap replicates
## 
## CALL : 
## boot.ci(boot.out = results, type = "bca", index = 3)
## 
## Intervals : 
## Level       BCa          
## 95%   (-0.1972,  0.2040 )  
## Calculations and Intervals on Original Scale

For climate enthusiasm

## BOOTSTRAP CONFIDENCE INTERVAL CALCULATIONS
## Based on 1000 bootstrap replicates
## 
## CALL : 
## boot.ci(boot.out = results, type = "bca", index = 4)
## 
## Intervals : 
## Level       BCa          
## 95%   ( 0.1850,  0.5687 )  
## Calculations and Intervals on Original Scale

For climate anxiety

## BOOTSTRAP CONFIDENCE INTERVAL CALCULATIONS
## Based on 1000 bootstrap replicates
## 
## CALL : 
## boot.ci(boot.out = results, type = "bca", index = 5)
## 
## Intervals : 
## Level       BCa          
## 95%   (-0.2876,  0.2340 )  
## Calculations and Intervals on Original Scale

For climate sorrow

## BOOTSTRAP CONFIDENCE INTERVAL CALCULATIONS
## Based on 1000 bootstrap replicates
## 
## CALL : 
## boot.ci(boot.out = results, type = "bca", index = 6)
## 
## Intervals : 
## Level       BCa          
## 95%   (-0.3272,  0.2545 )  
## Calculations and Intervals on Original Scale

For age

## BOOTSTRAP CONFIDENCE INTERVAL CALCULATIONS
## Based on 1000 bootstrap replicates
## 
## CALL : 
## boot.ci(boot.out = results, type = "bca", index = 7)
## 
## Intervals : 
## Level       BCa          
## 95%   ( 0.0102,  0.0259 )  
## Calculations and Intervals on Original Scale

For climate concern

## BOOTSTRAP CONFIDENCE INTERVAL CALCULATIONS
## Based on 1000 bootstrap replicates
## 
## CALL : 
## boot.ci(boot.out = results, type = "bca", index = 8)
## 
## Intervals : 
## Level       BCa          
## 95%   (-0.2773,  0.2750 )  
## Calculations and Intervals on Original Scale

Assumptions check

1. Distribution of the model residuals

Let’s have a look at the plots of model residuals:

2. Linear relationship between independent and dependent variables

3. No multicollinearity

We know from the correlation matrix that correlations of the independent variables do not exceed the customary cutoff point of .8 so we can say that this assumption is met. But let’s also have a look at the Variance Inflation Factor:

##      climate.anger   climate.contempt climate.enthusiasm    climate.anxiety     climate.sorrow                age 
##           3.328144           2.373049           1.627626           3.544578           3.418913           1.053609 
##         cc_concern 
##           4.040679

The value for VIF starts at 1 and has no upper limit. A general rule of thumb for interpreting VIFs is as follows:

  • A value of 1 indicates there is no correlation between a given predictor variable and any other predictor variables in the model.
  • A value between 1 and 5 indicates moderate correlation between a given predictor variable and other predictor variables in the model, but this is often not severe enough to require attention.
  • A value greater than 5 indicates potentially severe correlation between a given predictor variable and other predictor variables in the model. In this case, the coefficient estimates and p-values in the regression output are likely unreliable.
4. Homoscedasticity

Let’s check the plot of the predicted values against the standardized residual values from point 1 to confirm that the points are equally distributed across all the values of the independent variables.

Note

This HTML output presents the general logic of the analysis along with some results not outlined in the main body of the manuscript. Please note that the full R code for the data cleaning and data analysis is available in the supplementary materials on the accompanying OSF website.